Reliability Based Optimization
The traditional approach of optimisation, also known as deterministic optimisation (DO), does not account for any uncertainties in the system. Modern optimisation techniques, also known as probabilistic optimisation (PO), on the other hand, account for uncertainties that could affect the design objectives, such as uncertainties associated with mechanical properties or the geometry. Two well-stablished PO techniques are Robust-Design Optimisation (RDO) and Reliability-based Design Optimisation (RBDO). RDO focuses on minimising the sensitivity of the objective function to random changes in the uncertain variables in the system while RBDO aims at achieving a certain confidence in reliability of the product under a prescribed probabilistic constraint. In what follows, a brief literature review is conducted on different optimisation techniques.
Multi-objective reliability-based design optimisation results of the mono-stringer stiffened panel: (a) a large weight given to deterministic Nonlinear Post-buckling Strenght objective; (b) a large weight given to NPS with reliability of βt=5.0. For each case, squares and circles showing the pareto fronts of deterministic and reliable NPS, respectively.